We cloned the first human Smad gene (DPC4/MADH4/SMAD4) and found the first mutations of Smad genes in 1996. Since then, we were the first to identify human Smads 2 through 6, found the first mutations of Smad2 in cancer, developed the first knockout mouse model of Smad4, were first to define the DNA recognition element bound by Smads 3 and 4, developed the first SBE (Smad binding element)-based transcriptional reporters, proposed the co-crystal composition that allowed the first structural model of the MH1 Smad DNA-binding domain, published the first Smad4 MH1 domain structure/function analysis by scanning mutagenesis, produced a unifying theory of the functions affected by all DPC4 mutations known from human tumors, developed a nuclear import system for inducible Smad4 function, implicated MEK1 as a mediator of Smad-independent TGFB growth suppression, developed the first high-throughput compound screen for dissecting TGFB/Smad4 functions, identified the first examples in neoplasia of biallelic genetic inactivation of the TGFB type I receptor and of the activin type I receptor, developed using SAGE the first unbiased mammalian screen to identify Smad4 downstream targets, and validated a DPC4 assay for clinical tissue characterization. We propose to renew this highly productive effort for another five years. Our objective is an understanding of the role of Smad4-mediated signaling as a suppressor and genetic target in pancreatic tumorigenesis. Our long-term goal is the development of sensitive diagnostic techniques, preventative strategies for high-risk populations, and rational therapy for pancreatic cancer. In the proposed grant period, we will perform a set of essential studies that move us steadily toward these eventual goals: 1) Characterize mediators and defects in upstream Smad4 signaling. Recent discoveries suggest new directions for this work. 2) Identify downstream mediators of Smad4 suppressive functions. The suppressive functions of Smad signaling remain undefined, but would clarify the tie to tumorigenesis were they to be uncovered. 3) Perform compound screening to identify pathway-interacting agents. Such compounds immediately serve as tools for pathway dissection. Experience with such screening will also serve as a prelude to therapeutic targeting of Smad pathways in cancer.